Last month, I finished my PhD study 🎓@IllinoisCS and joined AI at Meta as a Research Scientist 🧐for #GeometricDeepLearning and #NetworkScience Wholeheartedly grateful for the support and understanding from my advisor, committee members, collaborators, friends, and my family ❤️

Three papers are accepted by #NeurIPS2024 🎉 1️⃣We investigate what Neural Tangent Kernel looks like on temporal graphs 2️⃣We investigate how to attack multivariate time series forecasting by GNNs 3️⃣ We investigate how to combine PageRank with Bandits to solve link predictions

We will present 6 papers at #NeurIPS2024 covering active learning, graph learning, time series and trustworthy ML. Come and have a chat during the poster sessions!🚀

Can we obtain temporal graph neural representation without neural network❓ Check out our #NeurIPS2024 paper, Temporal Graph Neural Tangent Kernel with Graphon-Guaranteed 🎉 Paper: openreview.net/pdf?id=266nH7k… Code: github.com/kthrn22/TempGN…

2 papers accepted by #ICLR2025, 1 paper by #AISTATS2025, and 1 paper by #KDD2025, congrats all collaborators 🎉 TL;DR 1️⃣ How to generate temporal heterogeneous graphs 2️⃣ How to tokenize a graph 3️⃣ How to infer invariant subgraphs 4️⃣ How to compress an evolving knowledge graph

This thursday, Feb 13th, 11am ET, at the Reading group, Katherine Tieu will present: Temporal Graph Neural Tangent Kernel with Graphon-Guaranteed (NeurIPS 2024). Looking forward to seeing you there! 🎉🎉 Zoom link on website openreview.net/forum?id=266nH…

Excited to know our #ICLR2025 paper is selected as Spotlight 🎉🎉🎉 We studied how to generate large-scale temporal heterogeneous graphs balancing privacy, utility, and efficiency We also found diffusion model may not be the solution to answer them all openreview.net/forum?id=tj5xJ…

💡 How can we describe a graph to LLMs ? 🧐 For example, G(n, p) uses number of nodes and connection probability to describe a graph. 📑 Please check out our survey, What Do LLMs Need to Understand Graphs: A Survey of Parametric Representation of Graphs arxiv.org/pdf/2410.12126

Beyond numerical-only time series models, let’s see how language can help time series forecasting and imputation🤹

🔥various graph types and various evaluation metrics

We'll present 4 papers and 1 keynote talk at #ICLR2025. Prof. Jingrui He and Prof. Hanghang Tong will be at the conference. Let's connect! ☕️

Identify Invariant Subgraphs for Spatial-Temporal OOD Problems 👏👏👏

Thrilled to have 3 #ICML and 1 #ACL accepted! Congrats all collaborators 🎉🎉🎉 Studying graph representation learning, multimodal alignment, and generative material design😄 Stay tuned🤔

Interesting to see the analysis between Inference scaling and LLM safety!

‼️We propose to learn temporal positional encodings for spatial-temporal graphs, L-STEP, in our #ICML2025 paper. Even simple MLPs can achieve leading performance in various temporal link prediction settings! 📄 Paper: arxiv.org/pdf/2506.08309 💻 Code: github.com/kthrn22/L-STEP